© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 5981
    Prev
    Next
    INDEX
    Explanations

    json structure keys

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron consistently lights up on specific named entities or rare technical terms—particularly proper names (e.g. color‐pigment names like “Manganese Violet” or “Blue 1 Lake”), user names, or other distinctive proper nouns.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dwóch
    0.57
     вторая
    0.54
     তড়িৎ
    0.54
     याता
    0.52
    sixth
    0.52
     दोन
    0.51
     ಮಿಶ್ರ
    0.51
    overset
    0.50
     trzech
    0.50
     crabs
    0.49
    POSITIVE LOGITS
    ↵↵
    0.71
    ↵↵↵
    0.70
    </code>
    0.59
    ↵↵↵↵
    0.57
    ↵↵↵↵↵
    0.56
    </blockquote>
    0.55
    ↵
    0.52
     promulg
    0.51
    .”
    0.51
    ))
    0.50
    Activations Density 0.001%

    No Known Activations